Canadian Wine's Share of LCBO Sales 80.0% 70.0% 60.0% Value Share 50.0% Volume Share 40.0% 30.0% Relative Price 20.0% 10.0% 0.0%

AMERICAN ASSOCIATION OF WINE ECONOMISTS

AAWE WORKING PAPER No. 191 Economics

THE COMPETITIVE POSITION OF ONTARIO’S WHITE TABLE WINES

Paul R. Masson

www.wine-economics.org Jan 2016

AAWE Working Papers are circulated for discussion and comment purposes. They have not been subject to a peer review process. The views expressed herein are those of the author(s) and do not necessarily reflect the views of the American Association of Wine Economists AAWE.

© 2016 by the author(s). All rights reserved. Short sections of text, not to exceed two paragraphs, may be quoted without explicit permission provided that full credit, including © notice, is given to the source. The Competitive Position of Ontario’s White Table Wines

Paul R. Masson, Weatherstone Consulting and University of Toronto1

Abstract

Canada has a small, but vibrant winemaking industry. Since the US-Canada Free Trade Agreement (FTA), which went into effect in 1987, growers have shifted to vinifera grapes and modern winemaking techniques and there has been an explosion in the number of wineries, which number about 150 in the Niagara peninsula alone. However, their share of the Ontario wine market, which fell with free trade, has continued to decline. This paper delves into the reasons for the downward trend.

Ontario provides a unique source of data on the wine market, since a single price for each wine is enforced by the Liquor Control Board of Ontario (LCBO). The paper analyses data downloaded from the LCBO’s website at the end of 2012 for all Ontario white table wines available for sale, and for wines from both “old world” (France and Italy) and “new world” (Argentina and Chile) wine producers. It is shown that after controlling for wine characteristics Ontario wines are higher priced than their competitors. This helps to explain why, despite improvements in the quality of Ontario wines, the share of imports in LCBO sales has risen. Certain wine varieties, in particular Chardonnay and Riesling, command higher prices, while in Vintages stores (but not in ordinary LCBO outlets), both the age of the wine and alcohol content have a significant positive effect on price. In terms of exports, Canada is miniscule on the world wine market, and it does not have a revealed comparative advantage in wine (except for ice wine).

In addition to the disadvantage on input costs, Ontario wine production also suffers from an industry structure that limits the extent that most wineries can exploit economies of scale. A limit in the number of off-winery stores--included in the FTA and subsequently NAFTA to prevent further protection of Canadian wines--has led to an uneven playing field in which two firms, one now American owned, operate the vast majority of those stores. Other wineries are limited to selling through the LCBO or at the winery. Further development of Ontario’s wine industry is likely to require opening up wine retailing to allow all wineries to benefit equally. In order to avoid the strictures of NAFTA, this would have to mean opening up competition to a wider range of wine retailers able to sell both domestic and imported wines.

1 Weatherstone Consulting, Box 1866, Niagara-on-the-Lake, Ontario L0S 1J0, Canada. Email [email protected]

Introduction Ontario’s wine industry, which is mainly based in the Niagara peninsula, has matured considerably since the time when Andres’ Baby Duck and Bright’s President Sherry were best sellers2. Until the 1980s, most of the grapes were of the Labrusca variety, including Concord and Catawba, which were hardy but made poor wine. In the face of the removal of protection as a result of the Free Trade Agreement (FTA) with the United States, Ontario grape growers introduced vinifera grapes and improved their grape-growing and wine-making techniques. As a result, Ontario wines no longer are scorned by wine connoisseurs at home and have achieved some recognition abroad. At present, there are more than 180 wineries in Ontario, and grapes are grown on 15,000 acres, making them the province’s largest crop in terms of farm gate value3. Quality of Canadian wines generally has been enhanced by a labeling system, Vintners’ Quality Alliance, which guarantees that a VQA wine is made from 100 percent Canadian grapes4. There are four Ontario wine appellations (Niagara, Pelee Island, Lake Erie North Shore and Prince Edward County) and a number of sub-appellations, describing the geographic origin of Ontario VQA wines.

Despite notable achievements, however, Ontario winemaking continues to face important challenges. Imported wine now constitutes more than two-thirds of the Ontario market; in contrast, Ontario wines were over half of domestic wine sales until the 1980s. Two factors clearly contributed to this trend: removal of protection (mentioned above), and the globalization of wine production and trade that has occurred over the past 3 or so decades5. The so-called “new world” producers, including notably Australia, New Zealand, Argentina and Chile, are now exporting large quantities of high-quality wines produced in climates more favourable to grape- growing than Ontario’s—at least for some varietals. Canada does not have an obvious comparative advantage in wine, except in one product, ice wine, for which Canada has established an international reputation. Overall, export sales have not grown to the same extent as increased wine imports. Even on the domestic market, it is not clear how Ontario’s table wines will maintain market share in the face of increasing global competition.

In this paper, trends in Ontario’s wine production and sales are described, presenting Ontario’s wine industry in a global context. Data from Ontario’s wine and spirits marketing board, the Liquor Control Board of Ontario (LCBO), are used to compare Ontario wine prices with those of major competitors. In doing so, I focus on white table wines—a market segment in which Ontario wines are traditionally the most competitive—and on wines from four competitors, two from the “old world” and two from the “new”— respectively France and Italy, and Argentina and Chile. Given its unique features and its market niche, ice wine is not included in the dataset, nor are French and Italian wines sold at the LCBO’s Vintages stores, which stock the rarer and

2 For a history of early Ontario winemaking, see Georges Masson (1979). John Schreiner (1984) gives a listing and evaluation of the Ontario wines available in the early 1980s. 3 Grape Growers of Ontario, http://www.grapegrowersofontario.com/ontarios-grape-and-wine-industry 4 See VQA Ontario, http://www.vqaontario.com/AboutVQA 5 See Anderson and Nelgen (2011) for a discussion of major trends.

2 more expensive wines, both imported and domestic6. However, their offerings at the regular LCBO stores are included.

The paper goes on to discuss the industry structure for wine-making in Canada. At present, there are two large Canadian (and Ontario) winemakers, namely Vincor (which was taken over by the US company Constellation Brands in 2006) and Andrew Peller Limited (which owns several wineries in Niagara, Nova Scotia, and in British Columbia). These companies have a special advantage in the industry, since they are allowed to operate off-winery stores in Ontario. In addition, they were instrumental in getting government approval for the “cellared in Canada” label, which allows them and other wineries to produce wine made from blends of domestic and imported grape juice, in undisclosed proportions. This wine labelling, combined with their marketing advantage which allows them to undercut higher quality VQA wines from smaller producers, limits further quality improvements for Canadian wines.

A concluding section speculates on the further evolution of Ontario’s wine making industry, which seems currently to be in a transitional stage. Ontario has not attracted the attention of oenologists and wine connoisseurs worldwide that other new wine producing regions have— partly, no doubt, because of its limited size. In addition, aside from ice wine, Canada is not known for a particular standout product, such as New Zealand Sauvignon Blanc, Australian Shiraz, or Argentina’s Malbec. The prices of Ontario wines are generally higher than those of imports, for similar quality. To achieve the size necessary for economies of scale so that Ontario’s wines can become more competitive, future consolidation of the industry is likely to be necessary. Greater competition, however, is likely to result not from the expansion of the existing two dominant producers, but from the creation of additional good-sized firms producing VQA wines. The role of the LCBO’s monopoly and the advantages of Vincor and Peller in marketing wine should therefore be re-examined in the context of making the industry more competitive7.

Trends in Ontario Wine Production and Consumption Because of its monopoly position in the sale and distribution of wine, the LCBO provides a wealth of data that account for virtually the totality of the wine market in Ontario. Not only are the aggregate volume and price data for all wine sold in Ontario published, the available selection of individual wines is easily accessed online as well as their price. The same is not true of other provinces or countries with more decentralized distribution, in particular where privately-run stores are allowed to purchase and sell wine freely. It is true, as already mentioned,

6 Canadian wines sold at Vintages are included, as are wines from Argentina and Chile. It is argued that given their long-established reputation and premium pricing, the Bordeaux classified growths and finer Burgundy wines of France, as well as the Barolos, Barbarescos, and Amarones of Italy, for instance, constitute market segments in which Ontario wines do not (at least at present) compete. 7 Policy issues that arise when considering reform of alcoholic beverage retailing are discussed in Masson and Sen (2014).

3 that a considerable amount of wine in Ontario is sold at the retail stores owned by Vincor (the Wine Rack, more than 160 stores) and Peller (the Wine Shop, 100 stores), as well as at individual winery stores. However, data for these sales are included in the LCBO totals. As for the list of wines offered, while the wines stocked at individual LCBO may differ, they are drawn from the same pool and are priced the same. Moreover, anyone can order from the LCBO’s wine list and have their selection shipped to their local wine store; winery sales must also be at the same price as in LCBO stores. Thus, as closely as is possible, Ontario’s wine market exhibits the “law of one price” for individual wines. In addition, since the LCBO prices the wines it purchases in order to clear the overall market, wine prices more nearly reflect underlying value in the province-wide market, given the preferences of consumers. For all these reasons the Ontario wine market provides an attractive source of data for econometric analysis.

To put Canada’s wine industry in perspective, Chart 1 gives trends in volumes of production, exports and imports. It can be seen that Canada’s production, after stagnating for two decades, has in the last decade risen modestly. Imports of wine, in contrast, have risen steadily since the end of Canada’s recession of the early 1990s. Canada’s exports, though they have shown a noticeable increase in the last 5 years, are still negligible when compared to the size of domestic production or of imports. Canada’s wine production is mainly accounted for by Ontario wines— 71 % of the total volume8--though of course the consumption of wine is much more evenly distributed across the country.

Chart 1

Canada Wine Production and Trade

400 350 300 250 imports 200 production 150 100 exports thousands of kilolitres 50 0

Source: Statistics Canada, Table 002-0010, and author’s calculations

8 Grape Growers of Ontario,

The next chart shows Canadian9 wine sales in the Ontario market relative to the size of the total market—including both sales at LCBO stores and those made at winery stores. These data show a continuing decline in both value and volume shares for Canadian wine, with the average price of Canadian wines relative to that of imports fluctuating around a fairly constant level of 80 percent. Thus, despite a well-recognized improvement in the quality of Ontario (and Canadian) wines, they have not been able either to increase market share (quite the contrary) or increase the relative price paid for Canadian wines. Of course, factors outside Canada, including marked quality improvements in other countries and the globalization of the wine trade are major factors explaining these trends.

Chart 2

100.0% 90.0% Canadian Wine's Share of LCBO Sales 80.0% 70.0% 60.0% value share 50.0% volume share 40.0% 30.0% relative price 20.0% 10.0% 0.0%

1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003 2006 2009 2012 Source: LCBO annual reports and author’s calculations

An indication of the global importance of a country’s wine industry, scaled by the country’s size, is suggested by the proportion of a country’s wine production that is exported. A ranking of wine-producing countries by Anderson and Nelgen (2011) indicates that in 2007-09 Canada was 19th of 23, ranking just above Uruguay, Georgia, Greece, and Brazil. Chile was first, Australia second, while Italy and France, large consuming as well as producing countries, ranked 8th and 11th, respectively.

A measure of whether a country has the climate, technology or resources to profit from the production of a particular good on world market is its “revealed comparative advantage,” defined as the share of a country’s exports of the good to its total exports, divided by the global export share of that good. A value in excess of unity indicates a comparative advantage, a value less

9 The LCBO does not split out Ontario wine sales from the total for Canadian wines. However, Ontario constitutes 71 % of Canadian production, and no doubt a higher percentage of the LCBO’s Canadian wine sales.

5 than unity the reverse. While it is difficult to get exactly comparable data, since the UN Comtrade statistics do not identify wine as a separate commodity, combining data from Statistics Canada for exports of grape wine and estimates by Rome’s Istituto di Servizi per il Mercato Agricolo e Alimentare10 of the value of world wine exports, we can calculate that the share of Canadian wine in Canada’s total exports is only about 0.0075 %, compared to a world share of wine exports of 0.17 %--yielding a revealed comparative advantage of only 0.045—far below the critical value of 1.0. Canada has a long way to go before it makes its mark on the world wine stage.

To put Canadian wine production in context, Table 1 presents the amounts produced by the most important wine-making countries. Canada’s production of about 1,600 thousand hecto-litres in 2012 (see Chart 1 above, noting difference in units) is dwarfed by France’s 41,400 thousand hecto-litres, as well as the output of other major European and new world producers.

Table 1. Global Wine Production (in thousands of hecto-litres) 2009 2010 2011 2012p France 46,361 45,704 49,633 41,400 Italy 47,450 48,525 41,580 40,100 Spain 35,166 35,235 34,300 30,000 other EU 33,921 26,912 31,371 EU total 162,898 156,376 156,884 USA 21,690 20,887 18,740 20,000 Argentina 12,135 16,250 15,473 11,800 Australia 11,710 11,240 11,010 12,700 Chile 10,093 9,152 10,572 12,600 non-EU 108,302 108,724 108,916 Global total 271,200 265,100 265,800 252,000 of which: Canada 1,390 1,349 1,465 1,587 p=preliminary Source: International Organization of Vine and Wine (OIV) , Cansim Table 002-0010, and author’s calculations.

Econometric analysis of Ontario wine prices Not surprisingly, given the small size of Ontario’s wine production in comparison to other producing countries, the liberalization of Canada’s trade, and increasing globalization of wine markets, the LCBO offers a wide selection of alternatives to Ontario wines. The catalogue of available wines is accessible online, and one can search for wines by grape type, country, region,

10 http://www.bloomberg.com/news/2012-04-24/world-wine-trade-increases-11-on-chinese-demand-ismea- says.html

6 or vineyard11. In order to compare Ontario wine prices with those of important competitor countries, data have been extracted from the LCBO’s wine list for all the Ontario wines listed (VQA only) and all listed wines from Argentina and Chile, plus the non-Vintages store offerings from France and Italy. The LCBO online catalogue reports several characteristics for each wine, in addition to whether the wine is available at the ordinary stores or at Vintages: price, year of vintage, producer name, country, region, contents of bottle, grape type (if specified), alcohol content, residual sugar, and sweetness rating. For some wines, a published quality rating by a wine critic may be given as well, but these are not widely available so they were not downloaded. In order to restrict the data to as homogeneous a sample as possible, only dry and semi-dry white table wines are included, leaving aside dessert wines such as ice wine; only VQA wines are included from among the Ontario wines, since they are guaranteed to be of Ontario origin. Some information will be provided later on the “cellared in Canada” offerings. Finally, only 750 millilitre bottles were included in the dataset.

Table 2 summarizes the data for the wines from Ontario and from the four competitor countries—398 observations in all, extracted from the LCBO’s online wine listing for end- November/early-December, 2012. Data on the residual sugar (in grams per litre) and on the sweetness rating were available only for some of the wines. While they were used as regressors for the wines for which they were available, they were never significant in explaining price. Only wines with a sweetness rating of 0 (extra dry), 1 (dry), and 2 (medium-sweet) were included in the dataset, so as to restrict the sample to table wines and exclude dessert wines (sweetness 3 or 4). The vintage of the wine was not always given on the LCBO’s website, but in some cases this seemed to be because both the most recent vintage and the previous year’s vintage were being sold at the same price. In our dataset, age was entered as 1 if either the vintage was specified as 2011 or was not reported; otherwise, age was calculated as 2012 minus the vintage year.

It can be seen that there is considerable variation in price, with Niagara wines having the highest average price at just over $20, while Italy’s were the cheapest (explained partly by the exclusion of Vintages wines for France and Italy). There are relatively few old bottles in the sample, though again this is due to exclusion of the Vintages offerings for France and Italy. The most common grape varieties are highlighted in the table (the classification actually included 20 different varietals, but the remaining ones were less frequent). Chardonnay is the commonest grape variety, followed by Riesling, Pinot Grigio (or in French, Pinot Gris), and Sauvignon Blanc; Torrontes is found only in Argentina’s wine in the sample. Of the 175 Niagara wines, 115 were found only in Vintages stores; the reason for this is primarily that the LCBO requires 200 cases of wine for listings in the regular stores, which many of the smaller Ontario vineyards cannot supply. Vintages sells wines available in smaller quantities, and hence is open to more Ontario wineries, some of which market their wines only at Vintages and at the winery store.

11 http://www.lcbo.ca/products/index.shtml

Table 2. Summary of LCBO Data on 750ml White Table Wines for Five Producers Variable No. of Mean Std. Dev. Min. Max. observations Argentina (59 wines) Price 59 15.04 5.76 6.95 39.95 Age 59 1.56 0.77 1 5 Alcohol 59 0.14 0.01 0.12 0.145 sugar (g/l) 41 4.24 2.07 2 11 sweetness no. (1-4) 56 0.23 0.47 0 2 Vintages? 41 chardonnay? 25 riesling? 1 sauvignon blanc? 0 torrontes? 20 pinot grigio/gris? 5 Chile (69 wines) Price 69 14.56 4.33 6.05 28.95 Age 69 1.61 1.03 1 5 Alcohol 69 0.13 0.01 0.12 0.148 sugar (g/l) 51 3.75 1.85 2 8 sweetness no. (1-4) 64 0.31 0.47 0 1 Vintages? 43 chardonnay? 27 riesling? 2 sauvignon blanc? 34 torrontes? 0 pinot grigio/gris? 0 France (49 wines) Price 49 13.89 4.04 8.45 26.5 Age 49 1 0 1 1 Alcohol 49 0.13 0.01 0.11 0.141 sugar (g/l) 49 5.86 5.89 2 25 sweetness no. (1-4) 49 0.27 0.57 0 2 Vintages? 0 chardonnay? 19 riesling? 1 sauvignon blanc? 9 torrontes? 0 pinot grigio/gris? 1 Italy (46 wines) Price 46 11.27 2.27 7 14.95 Age 46 1 0 1 1 Alcohol 46 0.12 0.01 0.11 0.132

sugar (g/l) 46 6.04 2.85 2 19 sweetness no. (1-4) 46 0.35 0.57 0 2 Vintages? 0 chardonnay? 6 riesling? 1 sauvignon blanc? 0 torrontes? 0 pinot grigio/gris? 21 Niagara (175 wines) Price 175 20.04 9.10 8.95 64.95 Age 175 2.18 1.46 1 8 Alcohol 175 0.12 0.01 0.084 0.145 sugar (g/l) 119 10.74 8.20 2 38 sweetness no. (1-4) 154 0.71 0.71 0 2 Vintages? 115 chardonnay? 64 riesling? 43 sauvignon blanc? 15 torrontes? 0 pinot grigio/gris? 8

Table 3. Regression of Price on Determinants Source Sum of Degrees of Mean Square Number of obs 369 squares freedom F( 13, 355) 33.18 Model 11925.22 13 917.32 Prob > F 0 Residual 9814.53 355 27.65 R-squared 0.5485 Adj R-squared 0.532 Total 21739.75 368 59.08 Root MSE 5.258

price Coef. Std. Error t P>t [95% Conf. Interval]

vintages 3.68 0.81 4.55 0.00 2.09 5.27 0 Age 2.94 0.31 9.39 0.00 2.32 3.55 0 alcohol 118.24 44.90 2.63 0.00 29.95 206.54 9 sweetno 0.39 0.51 0.76 0.45 -0.62 1.40 0 chardonn 1.95 0.80 2.44 0.01 0.38 3.53 ay 5 pgrigio 1.89 1.13 1.68 0.09 -0.33 4.11 5 riesling 1.29 1.16 1.11 0.26 -1.00 3.58 7 sauvblanc -0.17 1.06 -0.17 0.86 -2.26 1.91 9 torrontes -0.48 1.64 -0.29 0.77 -3.69 2.74 1 niagara 2.69 1.10 2.45 0.01 0.53 4.85 5 argentina -1.69 1.36 -1.25 0.21 -4.36 0.97 3 Chile -1.60 1.38 -1.16 0.24 -4.32 1.11 6 france 2.49 1.21 2.06 0.04 0.12 4.86 0 _cons -7.35 5.56 -1.32 0.18 -18.29 3.59 7

The regression results, presented in Table 3, are interesting on a number of counts. First, wines sold at Vintages, as expected, command a higher price, as do older wines and those with higher

10 alcoholic content. As mentioned above, the sweetness number (sweetno) does not come in significantly (its absence in some cases also shortened the sample to 369 wines). There is a significant difference in the price charged for different varietals: Chardonnay and Pinot Grigio on average command about $2 more, followed closely by Riesling, while Sauvignon Blanc and Torrontes have slightly lower prices, but not significantly so. Finally, after controlling for the other determinants Niagara and French wines have significantly higher prices than those of Italy (whose dummy variable was omitted for reasons of collinearity), while Argentina and Chile have lower prices than Italy’s, but not significantly so.

In order to discover whether the results are different for Vintages and non-Vintages wines, the sample was accordingly split into two. Table 4 presents the results for regular LCBO stores, and Table 5 for Vintages listings. For the regular listings, neither age nor alcohol content has a significant effect on price. While Chardonnay still commands a premium price, the same is even more true of Pinot Grigio. Interestingly, both Niagara and French wines have a significantly higher price than the others, and the premium on Niagara wines at $2.90, is only slightly less than France’s.

Table 4. Regression of Price on Determinants, Regular Stores Only Source Sum of Degrees Mean Number of 199 of Square obs F( 12, 186) 4.98 squares freedom Model 427.97 12 35.66 Prob > F 0 Residual 1331.44 186 7.16 R-squared 0.2432 Adj R- 0.1944 squared Total 1759.41 198 8.89 Root MSE 2.6755

price Coef. Std. Error t P>t [95% Conf. Interval]

Age 0.56 0.65 0.85 0.396 -0.73 1.85 alcohol 11.25 30.72 0.37 0.715 -49.35 71.85 sweetno -0.09 0.36 -0.26 0.793 -0.80 0.61 chardonnay 0.98 0.54 1.82 0.070 -0.08 2.04 riesling 0.50 0.80 0.63 0.529 -1.07 2.07 pgrigio 1.75 0.65 2.68 0.008 0.46 3.03 sauvblanc 0.05 0.76 0.06 0.948 -1.46 1.56 torrontes 1.20 1.56 0.77 0.442 -1.88 4.28 niagara 2.90 0.63 4.62 0.000 1.66 4.14 argentina -0.79 0.88 -0.90 0.369 -2.51 0.94 Chile 0.36 0.90 0.40 0.690 -1.41 2.13 france 3.07 0.65 4.75 0.000 1.79 4.34 _cons 8.44 3.79 2.23 0.027 0.97 15.91

When the regression is done with Vintages listings only, Niagara wines are compared solely to those of Argentina and Chile (since we have omitted French and Italian wines listed at Vintages from our sample). Table 5 shows that the premium for Niagara wines is considerably higher, $6.26, and continues to be significant with a p-value of 0.001. Varietal differences do not provide a significant explanation of prices, but age and alcohol content do.

Table 5. Regression of Price on Determinants, Vintages Stores Only Source Sum of Degrees Mean Number of 170 of Square obs squares F( 10, 159) 11.28 freedom Model 5637.06 10 563.71 Prob > F 0 Residual 7949.37 159 50.00 R-squared 0.4149 Adj R- 0.3781 squared Total 13586.43 169 80.39 Root MSE 7.0708

price Coef. Std. Error t P>t [95% Conf. Interval]

Age 2.93 0.45 6.50 0.000 2.04 3.82 alcohol 266.86 94.69 2.82 0.005 79.86 453.87 sweetno 1.11 1.06 1.05 0.294 -0.98 3.21 chardonnay 2.41 1.70 1.42 0.159 -0.95 5.77 riesling 3.45 2.49 1.39 0.168 -1.46 8.35 pgrigio 1.36 3.47 0.39 0.695 -5.49 8.22 sauvblanc -0.38 2.08 -0.18 0.855 -4.50 3.73 torrontes -0.67 2.76 -0.24 0.808 -6.13 4.78 niagara 6.26 1.79 3.50 0.001 2.73 9.79 argentina 0.53 2.02 0.26 0.793 -3.46 4.52 _cons -26.67 12.99 -2.05 0.042 -52.33 -1.01

The Structure of Ontario’s Wine Market12 As mentioned, Ontario (among other provinces) restricts the sale of wine to provincial stores or stores associated with wineries. The LCBO’s wine sales in the fiscal year ending March, 2011, totalled $1.701 billion, or 134,109 kilolitres, while sales by Ontario wineries and winery stores totaled $222 million, or 20,349 kilolitres13. The share of wineries and winery stores in the latest year was thus about 13% by volume, a figure that has fluctuated between 10 and 20% since 1970.

12 Lack of competition in Ontario’s wine retailing and recommendations for reform are discussed further in Masson and Sen (2014). Earlier proposals for reform (such as BASR, 2005) have run into opposition from vested interests and have not been implemented by the government of Ontario. A limited opening up of the retail sector, aimed mainly at generating greater government revenue, is currently being considered. 13 LCBO Annual Report 2010-2011. Note that the Annual Report also quotes net sales (which exclude taxes levied on wine). The above dollar amount is based on the gross price.

Chart 3

Ontario Wine Sales

180,000 160,000 140,000 120,000 100,000 winery stores 80,000 60,000 LCBO sales 000's kilolitres 40,000 20,000 0

1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003 2006 2009 2012 Source: various LCBO annual reports

The LCBO operated 634 stores (and 219 agency stores that are housed in grocery supermarkets or convenience stores) in the province in 2014, while the two operators of winery associated stores, Vincor and Peller (Andres), had some 260 stores, offering a more restricted range of wines, limited to their own production. However, the “cellared in Canada” label opens the door to much greater retailing by Vincor, given that its owner, Constellation Brands, owns vineyards in Argentina and Chile as well as the United States, providing a ready source of unfermented grape juice for blends with Canadian grape juice. Its online Wine Rack website in February 2013 listed a number of different brands, none of which was a Canadian VQA wine. The brands listed include mainly “International-Canadian Blends” (ICB) of Canadian and foreign wines (see Table 6). This term is now used by Vincor in place of “cellared in Canada,” which was criticized as not making clear that there was foreign grape content. However, the Wine Rack website says that the offerings are “made from a blend of foreign and Canadian wines,” not that that the wine is made in Canada from a blend of foreign and Canadian juice, which the “cellared in Canada” label seems to imply. The extent of the Canadian grape content of these wines is also unclear, since proportions of domestic and foreign grapes are unspecified and the regulations require only minimum 25 % Canadian content in their “cellared in Canada” brands.

Table 6. Listing of Wine Rack ICB Wines

Brand Sources of wine Varietal price 1/ Cape One South Cabernet Merlot $10 Africa/Canada Sauvignon Blanc $10 Leroisier France/Canada Merlot $10 Unoaked $10 Chardonnay Mazia Italy/Canada Pinot Grigio $11.95 Sangiovese $11.95 Zone Argentina/Canada Malbec Shiraz $10 Torrontes $10 Chardonnay Oak Ranch USA/Canada Cabernet Sauvignon $11.95 Chardonnay $10 White Zinfandel $11.95 Caleta Chile/Canada Cabernet Merlot $9.95 Chardonnay $9.95 Wallaroo Australia/Canada Cabernet Sauvignon $11.95 Trail Chardonnay $11.95 Shiraz $11.95 1/ $10 price requires purchase of 2 bottles Source: http://www.winerack.com/home/, accessed February, 2013.

The prices charged for the blends are considerably below comparable VQA prices, and in addition, they include grapes not usually grown in Canada such as Sangiovese, Torrontes, and Shiraz. In fact, these blended wines compete with some of the offerings from Constellation Brands’ vineyards in those countries, such as Fuzion, produced in Argentina. However, since they are to some (unknown) extent blended with Canadian wines, they can be sold at the Wine Rack, giving them a wider exposure.

The position of Vincor and Andrew Peller is privileged, in that they were grandfathered in the Free Trade Agreement with the United States: they were allowed to keep their wine retailing operations, although new winemaking operations are only able to sell wine on their premises. The relevant paragraphs of the Canada-US Free Trade Agreement are Article 804, para. 2 (these provisions were carried over to 312.2 section A of NAFTA):

“Notwithstanding paragraph 1, and provided that distribution measures otherwise ensure conformity with Chapter Five, a Party may: a) maintain or introduce a measure limiting on-premise sales by a winery or

distillery to those wines or distilled spirits produced on its premises; or b) maintain a measure requiring private wine store outlets in existence on October 4, 1987 in the provinces of Ontario and British Columbia to discriminate in favour of wine of those provinces to a degree no greater than the discrimination required by such existing measure. “

What is ironic is that the provisions of the treaty, which allowed for the maintenance of existing discrimination in Ontario and BC in favour of local wines, now serve to give a marketing advantage for the international blends of Constellation Brands. New measures, however, fall afoul of the free trade regulations, and a proposal by then Ontario PC leader Tim Hudak to improve the marketing of purely Canadian (VQA) wines, if implemented, would reportedly be challenged under the terms of that agreement by California winegrowers14. Thus, the restrictions on the sales of wine in the province, including LCBO regulations and the strictures of the free trade agreement, have the perverse effect of limiting the competitiveness of wines made from Ontario grapes in the domestic market, not protecting it.

Conclusions: What Future for Ontario Wines? Ontario winemaking has made enormous strides during the past three decades in the quality of the wines it produces and as a result, has earned a domestic and international reputation for Canadian wines. However, Ontario (and Canada) is a small wine producing region in a global context of generalized oversupply and fierce competition in all market segments—ordinary, premium, and super-premium wines. That competition is likely to get even fiercer as time goes on, given the capacity of newer wine exporting regions, especially Argentina, Chile, and South Africa, to increase supply and quality. Ontario wine faces stiff competition in all market segments. It has virtually no wine selling for under $10 a bottle, a segment in which Argentina, Chile, Italy, and South Africa have a number of LCBO listings. In the wines selling between $10 and $49 (which some classify as “premium” wines) there are many excellent offerings from the world’s major wine exporting countries that sell below Ontario wine prices, while Ontario wine has few wines (and available only in small quantities) that can claim “super-premium” status (over $50). As argued above, its wine prices are typically higher than other countries’ wines with similar characteristics. It has no uncontested advantages in terms of climate (except for ice wine) or terroir. Ontario winemakers must therefore strive to continue to improve quality while reducing costs relative to competitors in order to survive.

Further development of the Ontario wine industry is likely to require some easing of the constraints under which wineries operate and market their wines. While the LCBO provides a vehicle for marketing wines in moderate to large quantities, it does not provide the flexibility to suit all wineries, given its requirement that wineries deliver a minimum number of cases. The

14 See http://winesinniagara.com/2011/07/california-wants-to-take-on-the-ontario-wine-industry-what-a-pile-of- hooey/

16 alternative is selling at the winery, but this is not an efficient means of distribution because it requires the customer to travel to the production location. In contrast, the two grandfathered wine conglomerates, Vincor and Peller, profit from distribution networks with multiple stores throughout the province, often located near or in supermarkets where they are most convenient for consumers.

Unfortunately, giving all wineries the ability to have their own off-premises stores is impossible given current free trade agreements with the United States, which were designed to prevent Ontario’s provincial semi-monopoly on alcohol sales from giving Ontario wines an unfair advantage over US wines. The solution, given that the Ontario wine industry has now matured to an extent that it has been exposed to foreign competition for several decades, would be to go to full liberalization of wine trade—that is, getting rid of the LCBO’s monopoly, opening wine sales to private stores without restrictions (except health related), and putting the sale and distribution of all wines—domestic and foreign—on the same footing.

Such a reform could be expected to expand the market for wine—including Ontario wines—in the province by increasing the number of retail outlets and increasing the selection of wines locally available. It should stimulate competition among Canadian wineries by creating a more nearly level playing field among them, and provide an incentive for smaller wineries to further improve the quality of their products. By removing the advantages to the two major wine companies, Vincor and Peller, it might favour the creation of other middle-sized or large companies that could exploit some of the economies of scale of wine production and distribution. By lowering their costs, this would help to make VQA wines more competitive with imported wines, or international-Canadian blended wines.

References Anderson, Kym and Signe Nelgen (2011), “Wine’s Globalization: New Opportunities, New Challenges,” University of Adelaide, Australia, Wine Economics Research Centre Working Paper No. 0111, June.

BASR (2005). “A Strategy for Transforming Ontario’s Beverage Alcohol System. A Report of the Beverage Alcohol System Review Panel,” July.

Masson, Georges (1979). Wine from Ontario Grapes: A Guide to Winemaking with the New Hybrids, Niagara-on-the-Lake.

Masson, Paul R. and Anindya Sen (2014). “Uncorking a Strange Brew: The NEED FOR More Competition in Ontario’s Alcoholic Beverage Retailing System”, Commentary No. 414, CD Howe Institute, August 2014.

Schreiner, John (1984). The World of Canadian Wine, Vancouver: Douglas and McIntyre.